Is SmolLM3-3B or Qwen3-8B smaller?

SmolLM3-3B is smaller at Q4_K_M: 1.78 GiB against 4.68 GiB.

Which uses less memory at long context?

SmolLM3-3B — its KV cache at 32K context is 2.0x smaller, because of how its attention layers cache.

Model comparison

SmolLM3-3B vs Qwen3-8B

At Q4_K_M, SmolLM3-3B is the smaller download — 1,915,305,312 bytes against 5,027,783,488. At long context the gap widens: SmolLM3-3B's KV cache at 32K is 2.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	SmolLM3-3B	Qwen3-8B
Parameters	3.1B	8.2B
Architecture	smollm3	qwen3
Layers	36	36
Native context	65,536	40,960
Mixture of experts	no	no
Quantizations published	44	51
Smallest quantization	0.89 GiB	2.12 GiB
Q4_K_M	1.78 GiB	4.68 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	SmolLM3-3B	Qwen3-8B	Ratio
4,096	0.28 GiB	0.56 GiB	2.00×
8,192	0.56 GiB	1.13 GiB	2.00×
16,384	1.13 GiB	2.25 GiB	2.00×
32,768	2.25 GiB	4.50 GiB	2.00×
65,536	4.50 GiB	9.00 GiB	2.00×
131,072	9.00 GiB	18.00 GiB	2.00×